Disc drives are digital data storage devices which enable users of computer systems to store and retrieve large amounts of data in a fast and efficient manner. Disc drives of the present generation have data storage capacities in excess of several gigabytes (GB) and can transfer data at sustained rates of several megabytes (MB) per second.
A typical disc drive is provided with a plurality of magnetic recording discs which are mounted to a rotatable hub of a spindle motor for rotation at a constant, high speed. An array of read/write heads are disposed adjacent surfaces of the discs to transfer data between the discs and a host computer. The heads are radially positioned over the discs by a closed loop, digital servo system, and are caused to fly proximate the surfaces of the discs upon air bearings established by air currents set up by the high speed rotation of the discs. A plurality of nominally concentric tracks are defined on each disc surface to accommodate the storage of user data.
A preamplifier/driver circuit (preamp) generates write currents that are used by the head to selectively magnetize the tracks during a data write operation. The preamp further amplifies read signals detected by the head during a data read operation. A read/write channel and interface circuit are operably connected to the preamp to transfer the data between the discs and the host computer.
A rigid housing is provided to support the spindle motor and the actuator and to form an internal controlled environment to minimize particulate contamination of the discs and heads. A printed circuit board is mounted to the exterior of the housing to accommodate the disc drive control electronics (including the aforementioned servo circuit, read/write channel and interface circuit).
Disc drives are often used in a stand-alone fashion, such as in a typical personal computer (PC) or portable data processing/communication device where a single disc drive is utilized as the primary data storage peripheral. However, in applications requiring vast amounts of data storage capacity or high input/output (I/O) bandwidth, a plurality of drives can be arranged into a multi-drive array, sometimes referred to as a RAID ("Redundant Array of Inexpensive Discs"; also "Redundant Array of Independent Discs"). A seminal article proposing various RAID architectures was published in 1987 by Patterson et al., entitled "A Case for Redundant Arrays of Inexpensive Discs (RAID)", Report No. UCB/CSD 87/391, December 1987, Computer Science Division (EECS), University of California, Berkeley, Calif.
Since their introduction, RAIDs have found widespread use in a variety of applications requiring significant data transfer and storage capacities. It is presently common to incorporate several tens, if not hundreds, of drives into a single RAID. While advantageously facilitating generation of large scale data storage systems, however, the coupling of multiple drives within the same enclosure can also set up undesirable vibrations from excitation sources within the drives, such as spindle motors used to rotate the discs and actuators used to move the heads to various tracks on the discs. Such vibrations can be transmitted from drive to drive through chassis mounts used to secure the drives within the enclosure.
Vibrational components can be characterized as translational, or rotational in nature. Translational vibrations tend to move a disc drive housing back and forth along a plane of the drive, whereas rotational vibrations tend to rotate a disc drive housing about an axis normal to a plane of the drive. Because attempts are made to provide nominally balanced actuators, translational vibrations will generally have little effect upon the ability of the actuator to maintain the heads at a selected position with respect to the discs, as the discs and the actuator will both respond to the movement induced by such translational vibrations.
However, such is not usually true with rotational vibrations. Even with a nominally balanced actuator, rotational vibrations will tend to move the discs relative to the actuator because the actuator, acting as a free body, remains essentially undisturbed due to inertial effects while the discs, mounted to the housing, are displaced by imparted rotational vibration. When sufficiently severe, such movement will cause an "off-track" condition whereby a head is moved away from a selected track being followed. Such off-track conditions can adversely affect the ability of the drive to transfer data between the discs and host device.
The problems associated with rotational vibrations are well known in the disc drive art. Compensation attempts have included use of sensors that can detect the presence of rotational vibration in a disc drive, such as discussed in U.S. Pat. No. 5,235,472 issued Aug. 10, 1993 to Smith, assigned to the assignee of the present invention. Efforts to both detect and compensate rotational vibration using feedforward control include discussions by White and Tomizuka, "Increased Disturbance Rejection in Magnetic Disk Drives by Acceleration Feedforward Control," and Abramovitch, "Rejecting Rotational Disturbances on Small Disk Drives Using Rotational Accelerometers." Both of these papers were presented at the 13.sup.th Triennial World Congress, San Francisco, U.S.A., 1996.
While operative, there are limitations with these and other prior art approaches to minimizing the effects of rotational vibration in a disc drive. Sensors that specifically detect rotational vibration are commercially available, but are often prohibitively expensive for use in low cost disc drive designs and are also often difficult to properly calibrate for a particular drive application. Such sensors may include a piezoelectric polymer film disposed between metallic layers that detects rotational vibration in response to torsion induced on the film, as disclosed by the aforementioned Smith U.S. Pat. No. 5,235,472 patent; another construction uses multiple piezoelectric transducers within a single component enclosure to detect rotation in relation to differences in detected motion among the transducers.
Alternatively, rotational sensors can be formed from two or more discrete linear accelerometers which detect rotational vibration in response to differences in the detected motion between the devices. While potentially less expensive to implement than an integrated rotational sensor, commercially available discrete linear accelerometers (piezo or similar construction) can have significant part-topart output gain variation characteristics, making such unsuitable for use in a drive to detect rotational vibration without special screening and trimming operations to obtain matched sets of accelerometers.
By way of example, the aforementioned White et al. and Abramovitch references are illustrative of conventional approaches requiring use of relatively precise (and therefore expensive) accelerometers, as well as a calibration routine requiring use of a shaker table to impart vibrations of known characteristics. Such considerations make these approaches undesirable for high volume disc drive manufacturing, and prevent future adaptation of the response characteristics of a given drive to its subsequent field environment.
These references are also limited to compensating for rotational effects and do not address translational effects. However, translational effects have also been found to contribute to off-track errors due to actuator imbalance. In practice, induced vibration is seldom purely rotational or translational, but rather usually includes a combination of both.
Accordingly, as disc drive track densities and performance requirements continue to increase, there remains a continual need for improved approaches in the art to compensating for the effects of vibration in a disc drive using inexpensive and easily configured vibration sensor circuitry.